The Southeastern United States has the fastest growing Latino population in the country and at the same time carries a disproportionate HIV/AIDS disease burden. We propose a 2-year exploratory study (R21) to understand and characterize social and sexual networking patterns and socio-cultural and psychological correlates of sexual risk within a population about which little is currently known: newly-arrived, less-acculturated Latino gay men and Latino men who have sex with men (MSM) living in the rural Southeast. The AIDS case rate for Latino men is over 3 times that for non-Latino white men. The most common means of HIV transmission among Latinos is through same-sex sexual behavior. Among Latino gay men and Latino MSM in North Carolina, condom use rates are considerably lower than rates reported in other parts of the US. In a sample of MSM recruited in 3 bars in central NC, 69% of Spanish-preferring Latino MSM who reported having anal intercourse within the past 3 months reported inconsistent condom use with non-steady male partners. We also found that 32% reported having sex with both men and women within the past 3 months, and none of the men reported having ever been tested for HIV. Little is known about these men's social and sexual networking patterns and risk behaviors. This study will generate scientific knowledge and understanding needed to characterize risk and help identify potentially effective intervention approaches and strategies to reduce HIV exposure and transmission within this population. We will use 3 modes of data collection: (a) 30 rural Latino gay men and Latino MSM will complete iterative in-depth interviews (each respondent will be interviewed 3 times) to gain "insider" perspectives on HIV risk and prevention; (b) 16 local Latino-service and health providers will complete one-time in-depth interviews to gain "outsider" perspectives on HIV risk and prevention; and (c) 211 rural Latino gay men and Latino MSM will complete an interviewer- administered assessment to be developed based on formative interview data, through respondent-driven sampling (RDS). RDS is an innovative statistical approach to sampling to provide unbiased population estimates when a predetermined sampling frame is unknown. The proposed study will advance the field of HIV prevention research by generating in-depth knowledge necessary to develop culturally-appropriate models to better explain condom use within vulnerable communities: recently arrived, less-acculturated Latino gay men and Latino MSM who live in the rural Southeast. Study findings also will help shape new multilevel intervention approaches and strategies to reduce the disproportionate HIV/AIDS disease burden of Latino men. [unreadable] [unreadable] [unreadable]